Media generation
Everything an agent can create as media: images, short video with native audio, speech, transcripts, avatar videos, and maps.
Product shots for a launch, a narrated walkthrough, a transcript of yesterday's call, a map of the route. The media tools generate and edit that work directly in the thread, and everything made here becomes an artifact in the thread and your Library, ready to reuse or share.
Generated, not found
These tools create new media from your request. Finding existing images on the web is a separate always-available capability, and publishing pages and decks is Webpages and slides.
What an agent can make
Images, new or edited
Generate images from a prompt or edit ones you provide. Model, resolution, and aspect-ratio choices depend on the generation mode.
Short video with audio
Short clips with native audio. Duration, first-frame support, and reference controls vary by model.
Speech and dialogue
A single narrated voice or a multi-speaker exchange with a distinct voice per speaker.
A transcript of a recording
Speaker labels, per-segment timestamps, and detected emotion, so a meeting or interview becomes searchable and quotable.
An avatar reading a script
A talking-head presenter for a spokesperson clip, explainer, or personalized message, from HeyGen avatars and voices.
Maps and street views
Geocoding, places, directions, distances, an interactive map, and a Street View image of a real location.
Media generation tools
| Media | Powered by | Capabilities | Limits / specs |
|---|---|---|---|
| Images | Nano Banana Nano Banana Pro GPT Image 2.5 | Generates one image or a set in parallel. Edits images you provide by restyling or combining them, or by creating variations for product shots, hero banners, and redraws. | Resolution: 1K, 2K, or 4K. Aspect ratios: Square, 16:9, 9:16, and wider or taller formats. Model choice: Nano Banana for speed; Nano Banana Pro for fidelity; GPT Image 2.5 for crisp text rendering, dense compositions, and complex multi-image edits. |
| Video | Google Veo 3.1 FLUX 3 Gemini Omni 1.1 | Text-to-video, image-to-video, reference guidance, scene continuation, and native audio across three models. Veo 3.1 provides reference-guided generation. FLUX 3 delivers cinematic photographic fidelity with native audio. Gemini Omni 1.1 powers conversational video with synchronized dialogue and extensions. | Duration, resolution, and frame controls vary by model. See Video limits and specs below. |
| Audio | Google Gemini 3.8 Flash-Lite Text-to-Speech | Single-speaker narration or multi-speaker dialogue with a distinct voice for each speaker. 40% faster generation latency and native coverage across more than 100 languages. | Per call: Around 40 dialogue turns or roughly eight minutes. Longer audio: Generate in parts and stitch them together. |
| Transcription | Google Gemini multimodal transcription | Structured transcripts with speaker identification, per-segment timestamps, language detection, and detected emotion. | Formats: MP3, WAV, AAC, OGG, FLAC, and AIFF. File size: Up to 20 MB. Not supported: M4A. |
| Avatar video | HeyGen | Talking-head spokesperson clips, explainers, and personalized messages using avatars and voices from the connected HeyGen account. | Processing: Asynchronous; takes a few minutes. |
| Maps | Google Maps Platform | Geocoding, Places search, directions, distance calculations, interactive maps, Street View, weather, time zones, markers, routes, and street, satellite, or dark map styles. | Aerial View coverage: US addresses. Render time: One to three hours for an uncached cinematic 3D view. |
Video limits and specs
All three models support native audio and 16:9 or 9:16 aspect ratios.
| Model | Clip length | Resolution | Frame and reference controls |
|---|---|---|---|
| Google Veo 3.1 | 4–8 seconds | 720p or 1080p | First frame and up to 3 reference images |
| FLUX 3 | 6 or 8 seconds | 720p | First or last frames |
| Gemini Omni 1.1 | 3–10 seconds | 360p up to 4K | First and last frames |
Video extension: Gemini Omni 1.1 supports in-thread extension up to 40 seconds.
Media is not versioned
Unlike webpages and slides, generated images, audio, and video keep no version history. Regenerating doesn't replace earlier generations, it simply adds more artifacts to your thread.